Nuclear magnetic resonance imaging (MRI) is now a major diagnostic tool. However it is still used primarily in a qualitative fashion. We believe that a thorough and potentially quantitative understanding of signal changes observed in clinical MRI requires knowledge of how the basic building blocks of tissues behave, i.e. single cells. It is only recently that nuclear magnetic resonance (NMR) microimaging has enabled the direct examination of large single cells. Using NMR microscopy the water T1, T2 and multicomponent apparent diffusion coefficient (ADC) will be measured in the cytoplasmic and nuclear compartments of perfused single neurons as a function of hyper and hypotonicity. To provide an extracellular compartment and a controlled simple system a red blood cell a host model will be developed, and to provide a closer link to in vivo brain tissue, similar measurements will be performed on perfused hippocampal rat and human brain slices. Taken together, these data will be used to develop a realistic mathematical model of tissues in collaboration with Dr Stanisz in Toronto. Additionally we will begin preliminary studies examining the effects of a variety of perturbations on single cells, blood cell ghosts and brain slices. We believe that a thorough understanding of the signal characteristics of single cells and brain slices will lead to an improved understanding of the signal changes observed in MRI, which in turn may help develop a more quantitative approach to clinical MRI. This is especially important now that multicomponent ADCs, using long diffusion times, have been measured in humans, providing a compelling need for the studies in this proposal. The new understanding will ultimately improve the diagnostic sensitivity and specificity of MRI.